Using CA-Markov Model to Predict Land Use/Land Cover Changes in Bayer and al-Bassit region, Latakia, Syria

Abstract

Recently, land use change models have become important tools to support the analysis of land use dynamics. This research was aimed at the evaluating and predicting the land use/land cover change dynamics in Bayer and al-Bassit region of northwestern Latakia, Syria. In this paper, we used Cellular Automata and Markov Chain models to predict the LULC changes that are likely to occur by 2030 in Bayer and al-Bassit region. Three Landsat images acquired in the years of 1992, 2005, and 2018 were classified using Maximum Likelihood Classification algorithm and used as the input data for CA-Markov models. Kappa index was used to validate the model, and the overall accuracy recorded 79.34%. Based on a transition area matrix and transition rules a LULC map for the year 2030 were predicted. Compared to the LULC status of the reference year 2018, a significant reduction is likely to occur in 2030 in the forest area. This reduction might be in favor of the growth of agricultural land and urban area. The result shows CA-Markov model ability to predict future LULC changes in Bayer and al-Bassit region, and its importance for planners and land use managers.